Question: What are the best retargeting tools you’re seeing right now?
Peng: The way retargeting works, you have a list of IDs that you know are valuable to you. You want to take that further and find lookalikes, users who fit the same profile. A lot of ad networks are starting to offer retargeting. Facebook and a lot of the larger platforms are offering it now. It’s becoming ubiquitous.
But retargeting is dependent on how much data you have. You need a lot of installs before you can leverage retargeting at scale. Most developers, they profile between one and five percent of their conversion. There’s not a ton of that. So the question is, do you want to spend your time acquiring 1,000 users or retargeting a smaller pool of users? It’s highly dependent on how well your game is monetized in the first place.
Question: Has anyone explored email retargeting? I haven’t found a solution there at all.
AI Weekly
The must-read newsletter for AI and Big Data industry written by Khari Johnson, Kyle Wiggers, and Seth Colaner.
Included with VentureBeat Insider and VentureBeat VIP memberships.
Birnbryer: We don’t do it directly. We know some people who do it. There are some pitfalls. There’s all sorts of do-not-mail lists, and you have to be careful. When you’re shopping at Nordstrom and they say, “Would you like us to email you a receipt?” and then they hit you up with some other emails, you’re getting emails directly from Nordstrom. That would be something that, for a retargeter, would be very cool for your own platform. But like James said, you have to have a large user base already to start doing that.
That seems like one where it would be a pretty straightforward metric, though. You have your game players. You have their email. You shoot them a message. How much are we engaging with our email on our phones? Probably quite a bit. But not as much as, say, getting a push notification within the app itself. If you get that email on your desktop in your office, cool. You saw it. But you’re not going to go right back in the game. If you have a retargeting campaign that’s specific to mobile, that you include deeply, that takes someone directly to the level that they were in, that’s hyper-effective.
Bruch: Some of the partners we work with have said they have trouble attributing the value. Is it the person who generated the install, or the person who created that retargeted user? I don’t know if that’s something you’ve experienced as well.
Peng: We have some experience with email retargeting. We just tend to focus on using push or just new sources. We don’t spend a lot of time with it.
Question: How does user acquisition work in markets that aren’t using Facebook or email as much, like China? How is user acquisition different there?
Peng: We could have a whole different panel just for that. If you want to target a specific country, know what the top channels are. Know what the top social networks are. You have to do your homework if you really want to be there. You have to know the specific channels that will succeed, what people actually use. If you’re serious about it, go there. Look around. Look at how people interact with their apps. Twitter is very concentrated in certain geos. Certain networks are very specific to other geos. If you want to get very niche, spend some time and talk to developers and networks that focus on those areas.
China is probably the toughest UA market out there. If you’re going to target China, I recommend partnering with someone who has a very strong foothold in Asia. They do business very differently than we do here. I’m sure most of you know they have more than 190 app stores now. They’re dominated by five, but you have to have relationships with each of these app stores. They all have their own reasons why they would promote one app versus another. They have different revenue shares.
In the U.S., the best way to get recognized is to have Apple or Google feature your app on their main page where everyone is going and you get downloads. The same thing holds true for China. If you get featured by these app stores, they’re going to drive a bunch of downloads. But how you get featured in China is by having a strong relationship with those app stores, or you pay them a bunch of money to do it. You’re skipping some of the UA channels in China and going right to the app store. It’s like gaming the Apple Store with an up front contract. “We’re going to game your store.” “Okay, cool. We’re going to get paid a lot when you do that.” Know someone there, someone who knows the space.
Question: Do you think that the ad markets will eventually become more transparent, showing us all the sub-publisher IDs? And if you do think that’s going to happen, who do you think will be the first one to lead that trend?
Birnbryer: Transparency is super important. We’ll give you sub IDs all day long. If you want them identified, though, that’s a different story. We’ll give you every single sub ID you want. We want that. We’re trying to do LTV optimization for you, and we want to know as much as possible.
Where we are in the space now, it’s an interesting evolution, where as much as people have very cool tech, most of these companies are service companies. A service company is successful based on the strength of its relationships and its ability to navigate the space for you. You put your trust in an investment advisor because he knows the best stocks to pick for you. You put your faith in people because they’ve proven that they know exactly what they’re doing and they can do it fast.
Transparency is important to an extent, but ultimately that’s not the key factor. Key factor is being able to track your spend and track your return. I don’t care where it’s going as long as I can say, “This user drove this much value and I spent this much.” If we have a good enough relationship with someone, we’ll open up the list. If you have a strong relationship, if you’ve proven that you’ll work with us and you have no need to undercut us – we’ll open the books for you. There’s nothing wrong with that once that level of trust is there. Transparency is all about trust.